Wednesday, May 29, 2013

Poking Holes in Oladipo with New Shot Defense Metrics?


Victor Oladipo may be the best wing defender in this draft, but he is not without weaknesses that NBA scouts/execs are noticing with the help of Vantage Stats. Crowding shooters and getting a hand up without fouling is vitally important to effective defense.  In fact, we have already shown that in certain situations it will drop FG% by a full 10%. Utilizing a few of Vantage's innovative shot defense metrics, let's take a look at 5 prospects that will be called upon to defend the wing in the NBA: Victor Oladipo, Ben McLemore, Otto Porter, Shabazz Muhammad and Michael Carter-Williams.

Contest+
Vantage tracks 6 levels of shot defense, including block, alter, and contest (defined as when the defender is within 3 feet of shooter and his hand is up).  Contest+ is the percentage of shots defended where player blocks to possession, blocks to opponent's possession, alters, or contests.

Points Allowed Per Shot & Fouls Per Shot
Number of points allowed per shot defended includes free throws resulting from fouls and thus penalizes a player with a high Fouls Per Shot number.

Shots Defended Per Chance
Number of shots defended per defensive chance measures defensive activity and therefore gives context to box score counting stats in which level of activity is key.

Overall FG Against %
FG% by opponents on all shots defended.

The Numbers


Here is what NBA teams utilizing Vantage are seeing in the numbers from Oladipo, McLemore, Porter, Muhammad, and Carter-Williams:









Analysis


The numbers seem to highlight some weaknesses for Oladipo in shot defense - his foul rate is high and he allowed shooters to go almost 39% against him.  As a result, his Points Allowed Per Shot was .897.  To provide further context, Tony Allen allows .894 Points Per Shot against NBA-level talent.  Giving up a higher number of points to college shooters does not bode well for a player touted as an NBA-ready defender. However, a mitigating factor is his high Shots Defended Per Chance number and his high help rate (only Otto Porter averaged more helps per chance).  Thus, a lot of his points allowed were not when guarding his primary target.

We can't let Oladipo off the hook completely though.  Watching the video of his non-contested defense shows him relying on his active hands too much in help rather than playing with his feet, merely waving as guys go by, and he needs to temper his aggressiveness (especially when tired) so that he gives up fewer good looks. Oladipo exhibits the capacity to become a good NBA defender, but he is not there yet.




Keep checking in or follow us on Twitter as we continue to introduce new statistics in the following 10 categories:

1) Scoring
2) Facilitation
3) Rebounding
4) Screening
5) Turnovers and Fouling
6) Shot Defense
7) Disruptions
8) On-Ball/Screen Defense:
  • Keep in Front % (KIF%)
  • Close Out Points Allowed
  • Points Allowed Per Screen
  • Effective Screen Defense Rate
9) Help/Double Team Defense
  • Helps Per 100 Chances
  • Double Teams Per 100 Chances
  • Points Allowed Per Help/Double Team
  • Effective Help/Double Team Rate
10) Movement and Involvement

8 comments:

  1. Where are the statistic's on these player's "helps?"

    ReplyDelete
  2. I'm not sure these are apples-to-apples comparisons. Take Carter-Williams for instance. Given the fact that he played on the defensive perimeter in Syracuse's zone, one would assume and expect that the vast majority of shots taken against him were from 15 feet and out. It should come as no surpise, then, that his FG% against and foul numbers would be low. Compare this to Oladipo, who regularly had one-on-one responsibilities against an opponent's best offesive player both on the perimeter and driving to the basket.

    It would also be interesting to see these numbers normalized for quality of opponent. As stated above, Oladipo was regularly tasked with defending an opponent's best offensive player. I can't say that I watched enough Kansas, Syracuse, Georgetown, or UCLA games to say that the same is true for the other players. If it is not the case for the other players, that might also go a long way in explaining the discrepancies.

    ReplyDelete
  3. You forgot to account for his steals, silly. Those basically equate to his opponent shooting 0%, and also lead to free points on fast breaks.

    ReplyDelete
    Replies
    1. As stated in the post, we only presented metrics in the "Shot Defense" category. This post was not intended to be a fulsome analysis of Oladipo as a defender as we didn't show his numbers in our other defensive categories (Disruptions, On-Ball/Screen Defense, Help/Double Team etc.)

      Delete
    2. It's an easy adjustment to make, though. For instance, according to basketball reference, Oladipo recorded a steal on 4.6% of possessions.

      38.7*[0.142/(0.142+0.046)] = 29.2% shooting allowed after adjusting for steals. This will also help the other players, in proportion to how many steals they record.

      The risk of applying huge pressure like Oladipo does is that players can get around him more easily, but the reward is the higher steal rate. It's not fair to neglect the second half of that equation and conclude that he's not all he's cracked up to be defensively.

      Delete
    3. His disruption numbers are actually a bit higher than that (since we track things such as offensive fouls taken, deflections leading to steals and times where his pressure led to a non-steal turnover).

      Delete
    4. Interesting. Even taken in that context these are pretty surprising numbers, though. It should be possible to make a 1-number defensive rating based on statistics like these, and I look forward to seeing those big-picture results if and when they become available. Thanks for responding to my comments!

      Delete
    5. Also, I meant to add this, it should be possible to take the season-average points per play scored by the players a defender is guarding, and calculate a sort of "strength of schedule" correction for that defender. To do a good job I realize you would have to make further adjustments beyond what I said just there, but I think it would be a worthwhile endeavor, especially because player like Oladipo who guard multiple positions inevitably take on a tougher assignment on average than most defenders.

      Thanks again for your time.

      Delete